What is A ReproHack?
ReproHacks are low pressure and give participants the opportunity to work with real-world research data. They will also be able to network and learn from colleagues with different research and software experience.
The aim isn’t to undermine or discredit researchers or their work. Those organising and participating in ReproHacks recognise that reproducibility is a beneficial scientific activity in itself. The feedback that the events generate help researchers to improve their research, how it’s presented and ultimately aiding reproducibility and accelerating research.
How Does it Work?
In the run up to the event researchers are encouraged to submit papers, code and data to a repository ready for reproduction. As N8 CIR’s current research themes are Digital Health and Digital Humanities we would be keen to hear from researchers in these areas.
You can submit a paper here: https://sheffield-university.shinyapps.io/n8cir-reprohacks/
When submitting a paper you will need to include the following information:
- Your name, the paper title and where it was published
- An explanation of why we should attempt to reproduce this paper
- URLs for the paper, data and code
- A list of programming skills that would be useful to reproduce the paper
On the day, participants will begin by agreeing on how to approach the papers under consideration. They will also discuss and outline strategies and infrastructure to help ensure the events themselves are reproducible in the future.
After that it’s all about reproducing papers! Whilst working with the code participants will encounter a range of issues that will enable them to practice their coding and analytical skills. They will be using real data rather than small-scale dummy data usually used for such training. This means they are likely to be exposed to new code, or ways of using existing code, that they can take back to their departments and research groups. This will hopefully enable them to further research and science across universities.
Keeping notes along the way will enable constructive feedback to be offered to the researchers and authors who submitted a paper to the event. This feedback will help researchers to understand how their work is used, and re-used, by the academic community and how easy their data and code is to work with. It is well worth emphasising that these events are not about undermining or discrediting researchers or their work.
Hopefully this feedback will help researchers to arrange their future papers in ways that aid reproducibility and wider replication.
ReproHacks are all about developing approaches that help everyone to conduct scientific research in a better, more accessible and reproducible way. In this way, the entire scientific community benefits.
Some other individual benefits for research software engineers and data scientists include:
- Practical experience of reproducibility
- Experience of working with other people’s code and data
- Helping to improve reproducibility in future projects
- Increase the opportunities for the reuse of reproducible experiments
- Experience of working with ‘real’ data
Researchers also receive specific benefits such as:
- Feedback on how your data and code is saved and presented
- Advice on how to improve reproducibility in future projects
If your institution isn’t hosting one, or the date isn't convenient, but you want to get involved, don’t worry, ALL ReproHacks are open to anyone from across the N8 Research Partnership.
Travel bursaries are also available to PGRs and university staff from Durham University, Lancaster University and the University of York. They will be offered on a first come, first served basis. Please e-mail firstname.lastname@example.org for details of how to apply.
If you want to read more about the last ReproHack that Anna organised you can visit the Software Sustainability Institute blog here: https://www.software.ac.uk/blog/2019-08-07-reprohacking-carpentryconnect-manchester-2019
For any other queries please visit the Contact page at: https://n8cir.org.uk/contact/